Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Artificial Intelligence in Educational Decision Sciences Journal of Vocational, Informatics and Computer Education Journal of Applied Artificial Intelligence in Education
Andi Baso Kaswar
Okayama University
Computer Science & IT Decision Sciences, Operations Research & Management Education Electrical & Electronics Engineering Engineering Social Sciences

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Comparison of Dataset Proportions in SVM and Random Forest Algorithms in Detecting Student Dependence on AI in Learning Sardar Faroq Ahmd Khan; Pramudya Asoka Syukur; Andi Baso Kaswar; Marwan Ramdhany Edy
Artificial Intelligence in Educational Decision Sciences Vol 1 No 1 (2026): Artificial Intelligence in Educational Decision Sciences
Publisher : PT. Academic Bright Collaboration

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.66053/aieds.v1i1.6

Abstract

Purpose – The rapid integration of artificial intelligence (AI) in education has raised concerns about excessive student dependence, potentially undermining critical thinking and learning autonomy. This study aims to identify the most effective machine learning algorithm for detecting AI dependency in learning activities and to examine the impact of training–testing data proportion on predictive performance.Methods - This study employs the CRISP-DM framework and applies two supervised classification algorithms, Random Forest and Support Vector Machine (SVM), to a synthetic dataset of 10,000 AI-assisted learning sessions. The target variable, perceived AI assistance level, was discretised into three categories (low, medium, and high). Model performance was evaluated under four dataset split scenarios (60:40, 70:30, 80:20, and 90:10) using accuracy, AUC, precision, recall, and F1-score.Findings - The results show that Random Forest consistently outperforms SVM across all dataset proportions and evaluation metrics. The highest performance was achieved by Random Forest with a 60:40 split, yielding an accuracy of 67.6% and an AUC of 80.8%. Although SVM demonstrated stable performance, it required larger training datasets and remained inferior to Random Forest.Research limitations - The use of synthetic data and limited behavioural features restricts the generalisability of the findings. The moderate accuracy indicates that AI dependency is a complex construct not fully captured by the current model. Originality - This study provides empirical evidence on the combined influence of algorithm selection and dataset proportion in detecting AI dependency, offering practical guidance for developing early-warning systems to support responsible AI use in education.
AI Hallucinations in AI-Assisted Educational Decision-Making and Academic Honesty Intentions Among Undergraduates Desitha Cahya; Putri Ramdani; Annajmi Rauf; Andi Baso Kaswar; M Miftach Fakhri
Journal of Applied Artificial Intelligence in Education Vol 1, No 2 (2026): January 2026
Publisher : Academic Bright Collaboration

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.66053/jaaie.v1i2.9

Abstract

Artificial Intelligence in Education (AIED) is increasingly used to improve learning efficiency, personalization, and academic productivity. however, persistent risks such as AI hallucinations, algorithmic bias, and limited transparency can undermine the reliability of AI outputs and create ethical vulnerabilities that threaten academic integrity. This study aims to examine how students’ perceptions of algorithmic bias, perceived transparency of AI systems, and digital literacy influence their intentions to behave honestly when using AI for academic purposes. A quantitative cross-sectional survey was administered to 97 undergraduate students with experience using generative AI tools, and the proposed relationships were tested using Partial Least Squares Structural Equation Modeling (PLS-SEM) with SmartPLS 4. The results indicate that algorithmic bias (β = 0.248; t = 2.420; p = 0.008), transparency (β = 0.188; t = 1.920; p < 0.001), and digital literacy (β = 0.499; t = 5.457; p = 0.027) each have positive and significant effects on honest behavior intentions, with digital literacy emerging as the strongest predictor. These findings imply that strengthening students’ digital literacy together with institutional efforts to promote transparent and fairness-aware AI use can reduce unethical practices and foster a more integrity-centered academic environment in AI-assisted learning, while also informing ethical behavior frameworks for AIED implementation in higher education.
The Role of Anthropomorphism in Shaping Students’ Emotional Attachment to AIED: A Triangular Theory of Love Approach Asmi Ulfiah; Al Haytsam Mappaita; Aprilianti Nirmala S; Pramudya Asoka Syukur; Andi Baso Kaswar; Riyama Ambarwati
Journal of Vocational, Informatics and Computer Education Vol 3, No 2 (2025): December 2025
Publisher : Academic Bright Collaboration

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.66053/voice.v3i2.263

Abstract

In the digital learning era, Artificial Intelligence in Education (AIED) functions not only as an academic support tool but is also becoming an object of emotional attachment among students. While such attachment may enhance learning motivation, it also raises concerns about emotional dependence and its implications for students’ social and emotional well-being. This study investigates the effects of commitment, enthusiasm, emotional closeness, and anthropomorphic perceptions on students’ emotional dependence on AIED. A quantitative cross-sectional survey was conducted with 109 university students in Makassar using a 1–5 Likert-scale questionnaire. Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The structural model explained 62.7% of the variance in emotional dependence on AI (R² = 0.627), indicating moderate to strong explanatory power. Emotional closeness (β = 0.324; t = 2.893; p = 0.004) and anthropomorphic perception (β = 0.440; t = 4.871; p < 0.001) significantly increased emotional dependence, whereas commitment to continued AI use (β = 0.092; t = 0.883; p = 0.377) and enthusiasm toward AI (β = 0.081; t = 0.901; p = 0.367) were not significant predictors. These findings suggest that emotional dependence is driven more by affective engagement and the perception of AI as socially human-like than by cognitive motivation or usage intention. AIED interaction therefore extends beyond functional support into a relational experience resembling interpersonal connection. Given the limited geographic scope, future studies should involve broader populations and employ mixed-method approaches to deepen understanding of emotional dynamics in AIED use.
Co-Authors Al Haytsam Mappaita Annajmi Rauf Aprilianti Nirmala S Asmi Ulfiah Desitha Cahya M Miftach Fakhri Marwan Ramdhany Edy Pramudya Asoka Syukur Putri Ramdani Riyama Ambarwati Sardar Faroq Ahmd Khan